Modern data centers are managed using system management software or software asset management systems. These may need a comprehensive knowledge of installed software of the managed computing systems. However, there may be circumstances that lead to a mismatch between software that may be registered in a software catalog, e.g. a registry, and software that is actually installed on a given computing system. In order to properly manage the computing systems, it may be required to have a good match between supposedly installed software and actually installed software on a given computing system.
Several approaches have been made for discovering installed software on a given computing system. For example, the document U.S. Pat. No. 8,539,445 B2 discloses a method and a corresponding system and computer usable-medium for discovering software products on a plurality of computers. The discovering methods can populate a software catalog without the manual intervention of an administrator. Such an approach would reduce the cost of producing and maintaining a comprehensive knowledge base, which contains definitions of software products and the related signature.
Another approach has been disclosed by document WO 2006/088706 A2. Here, a method and apparatus are disclosed for identifying and cataloging software assets contained in a networked computer. In accordance with one embodiment, there is a method provided comprising providing a data warehouse including a master catalog file. The master catalog file comprises a software items table of attributes, a software packages table of attributes, and a mapping rule for mapping certain software item entries to certain software package entries.
However, the problem still exists that a mismatch may occur between registered software and software that may actually be installed on a computing system.